Integration: Power Rule, Definite Integrals and the Trapezium Rule
Integration reverses differentiation: it finds areas under curves and recovers a function from its gradient. For the TMUA you integrate powers of x by hand, evaluate definite integrals exactly, combine ranges, and judge whether the trapezium rule over- or under-estimates.
Part of the TMUA syllabus — revision for the Test of Mathematics for University Admission (TMUA), the UAT-UK maths test used by Cambridge, Oxford, Imperial, UCL, LSE, Warwick and Durham. No calculator; multiple choice.
Key points
- Power rule: add one to the power and divide by the new power (integral xn dx = xn+1/(n+1) + C), valid for every rational n except n = -1, where the rule breaks down.
- Indefinite integrals always carry an arbitrary +C. To solve dy/dx = f(x) you integrate to get y = integral f(x) dx + C, then substitute a known point on the curve to pin down C.
- Simplify BEFORE integrating: expand brackets, split fractions term by term, and rewrite roots as fractional powers (√(x) = x1/2, 1/x2 = x-2) so each term is a single power of x.
- A definite integral equals the SIGNED area between the curve and the x-axis: regions below the axis contribute negatively, so for a total physical area you split at the roots and add the magnitudes.
- Fundamental Theorem of Calculus: differentiation and integration are inverse operations, so integral a to b f(x) dx = F(b) - F(a) for any antiderivative F.
- Combining ranges: integral a to b + integral b to c = integral a to c (contiguous limits join up), reversing the limits flips the sign, and integral a to a = 0.
- Trapezium rule with n strips of width h = (b-a)/n uses n+1 ordinates: A ~= (h/2)[first + last + 2*(sum of the middle ordinates)].
- Judging the estimate: where the curve is concave up (bends upward, second derivative > 0) the straight tops lie above it so the trapezium rule OVERESTIMATES; concave down gives an UNDERESTIMATE.
Diagram
Formulae
integral xn dx = xn+1/(n+1) + C Integrating any single power of x; requires n rational and n ≠ -1. Apply once the term is written as x to a power.
integral a to b f(x) dx = F(b) - F(a) Evaluating a definite integral or an area once you have an antiderivative F; substitute the top limit minus the bottom limit.
integral a to b f + integral b to c f = integral a to c f Joining two integrals that share the middle limit b into one over the whole contiguous range a to c (or splitting one the other way).
integral a to b f = - integral b to a f Swapping the limits of a definite integral; the value changes sign. Useful to reuse a result you already computed.
A ~= (h/2)[y0 + yn + 2(y1 + ⋯ + yn-1)], h = (b-a)/n Estimating the area under a curve from n+1 equally spaced ordinates when you cannot (or need not) integrate exactly.
dy/dx = f(x) ⇒ y = integral f(x) dx + C Solving a first-order differential equation of this form; find C by substituting a given point on the curve.
Definitions
- Indefinite integral
- The general antiderivative of a function, written with an arbitrary constant +C because differentiating any constant gives 0, so infinitely many functions share the same derivative.
- Definite integral
- A single number, integral a to b f(x) dx = F(b) - F(a), equal to the signed area between the curve y = f(x) and the x-axis from x = a to x = b.
- Fundamental Theorem of Calculus
- The statement that integration and differentiation are inverse processes, so a definite integral is evaluated by taking any antiderivative at the two limits and subtracting.
- Trapezium rule
- A method that estimates a definite integral by cutting [a, b] into equal-width strips and replacing the curve on each strip by a straight line, i.e. a trapezium.
Worked examples
Evaluate the definite integral of (x + 1)/√(x) with respect to x from x = 1 to x = 4, giving an exact answer.
- 1
Split into powers of x:
f = x1/2 + x-1/2 - 2
Integrate term by term:
F = (2/3)x3/2 + 2x1/2 - 3
Evaluate at the top limit:
F(4) = 16/3 + 4 = 28/3 - 4
Evaluate at the bottom limit:
F(1) = 2/3 + 2 = 8/3 - 5
Subtract (FTC):
answer = 28/3 - 8/3 = 20/3
Answer: 20/3
Use the trapezium rule with 4 strips to estimate the area under y = x2 + 1 from x = 0 to x = 4, and state whether it is an overestimate or an underestimate.
- 1
List the 5 ordinates (h = 1):
y = 1, 2, 5, 10, 17 - 2
Apply the rule:
A = (1/2)[1 + 17 + 2(2 + 5 + 10)] - 3
Simplify:
A = (1/2)(52) = 26 - 4
Check concavity:
y'' = 2 > 0, so the curve is concave up - 5
Concave up means each straight top sits above the curve, so the estimate is too large.
Answer: Estimate A = 26, an overestimate (true area = 76/3 ~= 25.3).
Common mistakes
- ×Forgetting the +C on an indefinite integral. Without it you cannot solve dy/dx = f(x), because there is no constant to fix using the given point.
- ×Using the power rule at n = -1: you cannot write integral x-1 dx = x0/0, since dividing by zero is undefined; the rule only applies for n ≠ -1.
- ×Integrating a product or quotient term by term without simplifying first. The integral of a product is NOT the product of the integrals, so expand or split into separate powers before integrating.
- ×Confusing a definite integral with total area: when part of the curve is below the x-axis its contribution is negative, so integrating straight through undercounts the area. Find the roots and integrate each piece separately.
- ×Mixing up the trapezium over/underestimate rule. Concave up (a smiley curve) always OVERestimates; concave down UNDERestimates.
No-calculator tips
- ✓Integrate in your head with 'add one to the power, divide by the new power', then sanity-check by differentiating your answer back to the original.
- ✓Keep every answer exact as a fraction or surd (e.g. 20/3, not 6.67); TMUA options are almost always exact, so a decimal wastes time and invites rounding slips.
- ✓For a 'total area' question, sketch the curve and find its roots first to see whether any part dips below the axis before you integrate.
- ✓Read off concavity from a quick sketch (or the sign of the second derivative) to call the trapezium over/underestimate instantly, without ever computing the true area.
- ✓Reuse work: reverse a sign or join contiguous limits to build the integral you need from one you have already found, rather than re-integrating from scratch.
Test yourself
Original practice questions, no calculator. Work each out before revealing the answer.
Q1.The region R is enclosed by the curve y = x2 - 2x and the x-axis, between x = 0 and x = 2. What is the area of R?
- A. 8/3
- B. -4/3
- C. 4/3
- D. 2/3
Show answer
Answer: C — 4/3
On 0 < x < 2 the curve lies below the x-axis, so the definite integral is [x3/3 - x2] from 0 to 2 = -4/3; the area is its modulus, 4/3. Choosing -4/3 is the classic trap of treating a signed integral as an area.
Q2.The function g is defined by g(x) = integral from 2 to x of (t2 + 1) dt. What is the value of g'(3)?
- A. 6
- B. 9
- C. 22/3
- D. 10
- E. 12
Show answer
Answer: D — 10
By the Fundamental Theorem of Calculus g'(x) = x2 + 1 (the integrand), so g'(3) = 9 + 1 = 10. The 6 comes from wrongly differentiating the integrand (2x at x=3), and 22/3 from computing g(3) itself instead of g'(3).
Q3.The trapezium rule with two strips of equal width is used to estimate integral from 0 to 2 of x2 dx (ordinates at x = 0, 1, 2). Which statement is correct?
- A. The estimate is 3, and it is an over-estimate.
- B. The estimate is 3, and it is an under-estimate.
- C. The estimate is 8/3, and it is exact.
- D. The estimate is 6, and it is an over-estimate.
- E. The estimate is 4, and it is an over-estimate.
Show answer
Answer: A — The estimate is 3, and it is an over-estimate.
Estimate = (h/2)[y0 + 2y1 + y2] = (1/2)(0 + 2*1 + 4) = 3, versus the true value 8/3 approx 2.67; since y = x2 is convex (curves upward) the trapezia lie above it, so it over-estimates. The 6 drops the h/2 factor and the 4 uses a single strip, ignoring the middle ordinate.
Read this topic in the official UAT-UK TMUA content specification →
Keep going
- See how often each topic appears across past TMUA papers → — this one is about 11.6% of all questions.
- Practise Paper-1-style questions with the Oxford MAT archive → — 2007 to 2025, the closest ancestor of TMUA Paper 1.